that's interesting. for pain point #1, i wonder if compiling the dependent modules, and thereafter deleting the original source files would have sped up the im

that's interesting. for pain point #1, i wonder if compiling the dependent modules, and thereafter deleting the original source files would have sped up the imports using nothing more than the current set of python features?